Artemis

Drugs&Diseases
Similarity Calculation

Home

Drug

Disease

Document

>>Sign Out

Drug similarity method description

ATCSim

ATCSim calculate drug-drug similarity based on ATC code.
Input parameters:

ATC code file;
drug_ATC codes file,each row of data is drug ID in drugbank and corresponding ATC codes,for example DB00030\tA10AB01\tA10AB03.

Output parameter：

the result is saved in the output file as matrix,the value represents drug-drug similarity.

CDKSim

CDKSim calculate drug-drug similarity based on chemical structues using the Chemistry Development Kit(CDK)
Input parameter:

drug_smiles file,each row of data is drug ID in drugbank and corresponding smile,for example DB09019\tCN(CC1=CC(Br)=CC(Br)=C1N)C1CCCCC1

Output parameter：

the result is saved in the output file as matrix,the value represents drug-drug similarity.

SmileSim

SmileSim calculate drug-drug similarity based on SMILES representation of drugs.
Input parameters:

drug_smiles file,each row of data is drug ID in drugbank and corresponding smile,for example CN(CC1=CC(Br)=CC(Br)=C1N)C1CCCCC1\tDB09019
select one from the set{"TFIDF","EDIT","NLCS","HLCS","TF"} as the attribute parameter to invoke different method calculating similarity.

Output parameter：

the result is saved in the output file as matrix,the value represents drug-drug similarity.

Targets

Targets calculate drug-drug similarity based on targets information.
Input parameters:

targets file obtained from drugbank
drug id list1,each row of data is drug ID in drugbank,the rowname is required in the first row
drug id list2 in drugbank

Output parameter：

the result is saved in the output file as matrix,the value represents drug-drug similarity.

DDIs

DDIs calculate drug-drug similarity based on drug interaction information.
Input parameters:

interactions file obtained from drugbank
drug id list1,each row of data is drug ID in drugbank,the rowname is required in the first row
drug id list2 in drugbank

Output parameter：

the result is saved in the output file as matrix,the value represents drug-drug similarity.

Disease similarity method description

BOG

BOG calculate disease similarity using the BOG method.
Input parameters:

disease id or disease name list1,for example Muscle Cramp
disease id or disease name list2,for example Muscle Cramp
disease_gene file,each row is disease\tgene,for example spinal cord diseases\t203228

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

CosineDFV

CosineDFV calculate cosine similarity of disease.
Input parameters:

disease id or disease name list1,for example Muscle Cramp
disease id or disease name list2,for example Muscle Cramp
file storing the relationship between disease and other biological entities,each row is biological entity\tdisease\tfrequency,for example Fever\tBorrelia Infections\t8

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

doSims

doSims Given two sets of DOID, this module can calculate the semantic similarity of diseases. The module uses 5 semantic similarity methods to calculate disease similarity.
Input parameters:

disease DOID list1,for example DOID:14095
disease DOID list2,for example DOID:14095
select one from the set{"Wang", "Resnik", "Rel", "Jiang", "Lin"} as the attribute param to invoke different methods.

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

FunSims

FunSims calculate disease similarity.
Input parameters:

disease id or disease name list1,for example Muscle Cramp
disease id or disease name list2,for example Muscle Cramp
disease_gene file,each row is disease\tgene,for example spinal cord diseases\t203228
file storing gene-gene relationship,obtained from HumanNet,each row is gene\tgene\tsimilarity_value,for example 10755\t4646\t0.95

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

Sun_topologys

Sun_topologys calculate disease similarity using Sun_topology method.
Input parameters:

disease id or disease name list1,for example Muscle Cramp
disease id or disease name list2,for example Muscle Cramp
disease_gene file,each row is disease\tgene,for example spinal cord diseases\t203228
gene_graphlet signature file,each row is gene\ta vector of 73 numbers. The graphlet signature of the gene is calculated based on the location of the gene in PPI.

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

Sun_functions

Sun_functions calculate disease similarity using Sun_function method.
Input parameters:

disease id or disease name list1,for example DOID:14095
disease id or disease name list2,for example DOID:14095
disease_GO file,each row is disease\tGO,for example DOID:5652\tGO:0006357

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

ICods

ICods calculate disease similarity using ICod method based on the relationship of disease and genes and gene/protein network data.
Input parameters:

disease id or disease name list1,for example Muscle Cramp
disease id or disease name list2,for example Muscle Cramp
disease_gene file,each row is disease\tgene,for example spinal cord diseases\t203228
gene-gene network file,each row is gene\tgene,for example ITGA7\tCHRNA1

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

PSBs

PSBs calculate disease similarity using PSB method based on the disease_GO relationship and GO_gene relationship
Input parameters:

disease id or disease name list1,for example Muscle Cramp
disease id or disease name list2,for example Muscle Cramp
disease_GO file,each row is disease\tGO,for example DOID:5652\tGO:0006357
GO_gene data set,each row is GO\tgene,for example GO:0006357\tBACH1

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

Separations

Separations calculate disease similarity using Separation method based on the relationship of disease and genes and gene/protein network data.
Input parameters:

disease id or disease name list1,for example Muscle Cramp
disease id or disease name list2,for example Muscle Cramp
disease_gene file,each row is disease\tgene,for example spinal cord diseases\t203228
gene-gene network data set,each row's format is gene\tgene.for example ITGA7\tCHRNA1

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

Sun_annotations

Sun_annotations calculate disease similarity using the Sun_annotations method.
Input parameters:

disease id or disease name list1,for example Muscle Cramp
disease id or disease name list2,for example Muscle Cramp
disease_gene file,each row is disease\tgene,for example spinal cord diseases\t203228

Output parameter：

the result is saved in the output file as matrix,the value represents disease-disease similarity.

Module specification

Attention:All modules can have multiple outputs, so the output of each module is encapsulated in dict.

1.Java

Each module contains only one file, the file type must be. java file and the file name must be consistent with the class name contained in the file;
The module needs to include the execute method (similar to the main method);
The execute method's input object type is InputStream;
The attribute parameters of execute method are string. If you need to convert them, please convert them into other types (such as int) in the module;
If the input object is a triplet file, attribute parameters should be added to indicate the number of columns to be read.
Execute method's output object type is Map < String, String > type, example: <'output 1','This is a string'>;

2.Python

Each module contains only one file, and the file type must be.Py file. Python3.x syntax is recommended;
Attribute parameters are string types.
If the input object is a triplet file, attribute parameters should be added to indicate the number of columns to be read.
Each module has one or more output parameters, specifying the result to write to the location of the file;
The result is stored in dict type ,converted to JSON format and then wrote to file.